This research will examine how "molecular" (small-scale) variables affect choice behavior. One line of research will examine how individuals choose between a small immediate reinforcer and larger delayed reinforcer, and how the strength or value of a reinforcer declines with increasing delay. These experiments will test different mathematical models of the temporal discounting function. A second line of research will examine different variables that may modulate the effects of delay, including the sizes of the reinforcers involved, an individual's past experience with delayed reinforcers, the amount of effort committed, and the presence of signals that partition the delay interval into smaller segments. An adjusting-delay procedure will be used, in which the delay for one alternative is systematically increased and decreased over trials to find an indifference point-a delay at which the two alternatives are equally preferred. A third line of research will examine choice in more complex procedures called concurrent-chains schedules, which feature a choice period followed by a period in which the subject receives the consequences of its choices. These studies will be designed to test and compare the predictions of several current mathematical models of complex choice. The subjects will be pigeons and rats performing in standard operant conditioning chambers with food as a reinforcer. In some cases, parallel experiments will be conducted with the two species to test the generality of the findings. Many important everyday choices involve a conflict between an individual's short-term and long-term interests (such as the pleasures of smoking, drinking, or overeating versus future health, or the fun of a shopping spree versus saving for long-term financial security). The research described in this proposal is directed at developing a quantitative model for understanding how such choices are made, why individuals sometimes do and sometimes do not make choices that are in their best long-term interests, and what factors control these decisions. [unreadable] [unreadable]